So, how does a Border poll get called?
The 1998 Belfast Agreement, also known as the Good Friday agreement, sets out in detail how relationships should be managed across these islands, as well as the rights and protections that should be afforded to different communities in Northern Ireland.
It makes it clear that the secretary of state for Northern Ireland, a UK government minister, holds the power to call a Border poll on Irish unification. However, the text of the agreement is vague – perhaps deliberately vague – on crucial aspects of how the power should be exercised.
Some of the rules in the agreement are straightforward. The Northern secretary can call a Border poll at any time if they wish, as long as one has not been held in the previous seven years.
READ MORE
More difficult, however, is the circumstance in which they must call a Border poll. A poll must be called, the agreement says, if at any point it “appears likely” to the Northern secretary that a majority of those voting in a Border poll in Northern Ireland would support unification.
The emphasis is on supporting change; voters appearing to be ambivalent, open to unity, or in favour of holding a referendum to reject unity at the ballot box is not sufficient. It needs to appear to the minister that there is a majority for changing the status quo and moving to a united Ireland.
This invites the obvious question: how would a secretary of state take this decision, and measure whether something “appears likely”? The agreement does not say.
The most obvious place to look is opinion polling. Different companies use different methodologies (face-to-face polling, online polling etc), leading to some variation in the results. Generally, however, polling in recent years has shown more support for retaining Northern Ireland within the UK than for leaving, with a modest but nonetheless noticeable uptick in favour of Irish unity since Brexit.
In most polls, those who “don’t know” or are open to change appear to hold significant sway. If they were persuaded that unity was the right course for Northern Ireland, they could tip the balance.
[ As momentum builds in the Irish unity debate, how is unionism responding?Opens in new window ]
Another way to assess what “appears likely” as to support for unification is by political party performance. Northern Ireland’s party system is very clearly shaped by views towards the constitutional question, with the Democratic Unionist Party (DUP) and smaller unionist parties such as the Ulster Unionist Party (UUP) and Traditional Unionist Voice (TUV) supporting Northern Ireland’s constitutional place in the union. Similarly, on the nationalist side, both Sinn Féin and the Social Democratic and Labour Party (SDLP) support moving towards a “new” or united Ireland. Closely watching election results is another option for the secretary of state.
Here, again, we find an even balance on support for unification. Much has been made of Sinn Féin becoming the single largest party overall and taking the first minister position from the DUP. The last Westminster election, in 2024, also led to half of Northern Ireland’s 18 seats being occupied by constitutional nationalists.
A more detailed look at that election, however, shows that unionism overall continues to command a plurality of the votes, though neither side has a majority of voters. As with polling, constitutionally non-aligned voters (supporting parties such as Alliance and the Greens) hold the balance in an increasingly fragmented political system.
Alongside polling, the Northern secretary might look at how many votes these different parties receive in future elections, or how many seats they win in the assembly or Westminster, when making up his or her mind about what “appears likely” when it comes to unification.
A future vote in the assembly could also reveal that at least half of MLAs were in favour of unification.
Finally, the secretary of state might look at other ways of measuring public opinion, such as the census or qualitative focus groups. Both could give insight into the demographic realities of the electorate, and what people think about politics when given more time and space to set out their views. However, neither of these is a good assessment of what “appears likely” that voters will do.
Since these various sources of evidence could produce quite a mixed picture, a secretary of state for Northern Ireland will probably want to use a variety of sources of evidence, over time, to get a rounded sense of what appears likely.
Ideally, the decision would be taken in close consultation with the government of Ireland too, as there would also need to be a referendum south of the Border. Moreover, all decisions by the UK government on this front in Northern Ireland’s case must be taken, as per the Belfast Agreement, with “rigorous impartiality”.
Many people, understandably, want the Northern secretary to set out clear criteria for making their decision now, in advance of any constitutional process. There has even been a court challenge on the question, though the courts said the Northern secretary did not have to set out criteria, and retained political flexibility on making this determination. A UK government might also be less inclined to set out clear criteria for when a Border poll would be held when public opinion in Scotland on independence is so finely balanced.
None of this is straightforward, legally or politically. What looks like good criteria for measuring public opinion now might not be useful in a few years. In an age of growing misinformation, will polling still be a reliable source of public opinion information in a decade? Will the party system in Northern Ireland remain unchanged, giving clear indications of the balance of power between unionism and nationalism, or will it be reshaped, as so many others have been across Europe in recent years?
It is clear there are ways of assessing whether the “appears likely” clause of the Belfast Agreement has been met. Equally, it is clear that the conditions for holding a referendum have not been met at this point in time. That said, across polling, elections, and other sources of information on the state of public opinion, it seems there is growing interest in discussing unification. If that interest is sustained, and support for unification grows with it, the secretary of state will come under increasing scrutiny as to how they decide when, if at all, to hold a Border poll on Irish unity.
Dr Conor J Kelly is the Bingham Early Career Fellow in Constitutional Studies at Balliol College, University of Oxford; Prof David Kenny is Professor in Law and Fellow at Trinity College Dublin
